© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13920
    Prev
    Next
    INDEX
    Explanations

    why questions

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is highly responsive to numeric values and statistical measurements (e.g., counts, percentages, p-values) in scientific text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    伀
    -0.91
    identiality
    -0.90
    )=>{
    -0.86
    Rispondi
    -0.82
    Juri
    -0.81
     gerçekleşti
    -0.80
    することで
    -0.79
    sikan
    -0.79
     aggressively
    -0.79
     will
    -0.79
    POSITIVE LOGITS
     Why
    1.46
     why
    1.45
    Why
    1.42
    why
    1.29
     successful
    1.18
     क्यों
    1.18
     Почему
    1.15
     despite
    1.13
     mengapa
    1.11
     seem
    1.09
    Activations Density 0.355%

    No Known Activations